Mean-Based Error Measures for Intermittent Demand Forecasting
TL;DR: This paper proposed several new error measures with wider applicability, and correct forecaster ranking on several intermittent demand patterns, called mean-based error measures, which evaluate forecasts against the (possibly time-dependent) mean of the underlying stochastic process instead of point demands.
read more
Abstract: To compare different forecasting methods on demand series we require an error measure. Many error measures have been proposed, but when demand is intermittent some become inapplicable, some give counter-intuitive results, and there is no agreement on which is best. We argue that almost all known measures rank forecasters incorrectly on intermittent demand series. We propose several new error measures with wider applicability, and correct forecaster ranking on several intermittent demand patterns. We call these "mean-based" error measures because they evaluate forecasts against the (possibly time-dependent) mean of the underlying stochastic process instead of point demands.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Evaluating predictive count data distributions in retail sales forecasting
TL;DR: In this article, the authors explore point forecast accuracy measures, explain why MAD, MASE and wMAPE are inherently unsuitable for count data, and use the randomized Probability Integral Transform (PIT) and proper scoring rules to compare the performances of multiple causal and non-causal forecasting models on two datasets of daily retail sales.
163
Nonstationary Z-score measures
TL;DR: This work develops advanced techniques for measuring bank insolvency risk and introduces novel estimators whose aim is to effectively capture nonstationary returns, and introduces bias reduction strategies for state-of-the-art Z-Score measures in the literature.
72
Forecast Information Sharing for Managing Supply Chains in the Big Data Era: Recent Development and Future Research
Bin Shen,Hau-Ling Chan +1 more
TL;DR: A comprehensive review on forecast information sharing for managing supply chain in the big data era is conducted and the value and obstacles of sharing forecast information are discussed.
60
Spare parts demand forecasting: a review on bootstrapping methods
TL;DR: This paper reviews as well the service level models associated with the bootstrapping approach with an emphasis on the fill rate models for a better understanding of its framework and retrieves more robust explanations of the observed mixed-performances of the available bootstrap-based forecasting methods.
58
A new method to forecast intermittent demand in the presence of inventory obsolescence
TL;DR: In periods with positive demand the new method updates the demand sizes, the demand intervals, and the estimator, similar to SBA, but in any time period if the actual demand interval becomes higher than the most recent estimated demand interval, the update becomes in every period similar to the probability of occurrence in TSB.
58
References
Another look at measures of forecast accuracy
Rob J. Hyndman,Anne B. Koehler +1 more
TL;DR: In this paper, the mean absolute scaled error (MESEME) was proposed as the standard measure for comparing forecast accuracy across multiple time series across different time series types, and was used in the M-competition as well as the M3competition.
5K
The M3-Competition: results, conclusions and implications
Spyros Makridakis,Michèle Hibon +1 more
TL;DR: In this paper, the M3-Competition, the latest edition of the M-Competitions, is described and its results and conclusions are compared with those of the previous two M-competitions as well as with other major empirical studies.
1.7K
The accuracy of extrapolation (time series) methods: Results of a forecasting competition
Spyros Makridakis,A. Andersen,A. Andersen,R. Carbone,Robert Fildes,Robert Fildes,Michèle Hibon,R. Lewandowski,J. Newton,E. Parzen,Robert L. Winkler +10 more
TL;DR: The results of a forecasting competition are presented to provide empirical evidence about differences found to exist among the various extrapolative (time series) methods used in the competition.
1.5K
Exponential smoothing: The state of the art
TL;DR: A critical review of exponential smoothing since the original work by Brown and Holt in the 1950s is shared, which concludes that the parameter ranges and starting values typically used in practice are arbitrary and may detract from accuracy.
1.5K
Error measures for generalizing about forecasting methods: Empirical comparisons
J. Scott Armstrong,Fred Collopy +1 more
TL;DR: In this article, the authors evaluated measures for making comparisons of errors across time series and found that the median absolute error of a given method to that from the random walk forecast is not reliable, and therefore inappropriate for comparing accuracy across series.
1.4K